Most traders put the stop where the loss stops hurting, which is exactly why it keeps getting hit. The market has no clue what your account balance is or what a $150 loss means to you. It goes where liquidity and structure send it, and a stop placed for comfort instead of for the chart is a donation with extra steps. This article covers the three honest ways to locate a stop (technical, ATR, and time), the arithmetic that turns a stop into a position size, and the funded-account rule that decides whether a trade fits at all.
The two jobs of a stop
A stop does two things at once, and mixing them up is the root of most bad placement. First, it invalidates the trade idea. It marks the price at which the reason you entered is objectively proven wrong. Second, it caps the loss. It defines how much money leaves the account if you are wrong. Those are not the same job. A good stop nails the first, and sizing takes care of the second.
The invalidation level is a fact about the chart. It sits where participants would have to drive price to prove your thesis dead, and it does not care how many dollars that represents. The dollar loss is a fact about your account, and it falls out of stop distance, contract multiplier, and size. You set all three at the sizing step. Keep the two apart and you have the whole discipline. Collapse them (place a stop at the price where the loss "starts to feel uncomfortable") and the market punishes that category error with routine, random stop-outs.
Invalidation is a fact about the chart. Comfort is a fact about your account. The stop handles the first, sizing handles the second.
Structural stops
A structural stop (some traders call it a technical stop) sits just beyond the price structure that defines your setup. For a long, that means below the swing low or below support. For a short, above the swing high or above resistance. Trading a range? It sits outside the range edge. The logic is plain. The market respects structure, so once price trades through the level that defined your thesis, participants have pushed past the point that made the trade make sense. The setup genuinely failed. You were not unlucky, you were wrong, which is a far more useful thing to be.
Add a small buffer beyond the exact level, a few ticks, so you sit behind the line rather than right on it. The level itself is where resting stops and liquidity pile up, and it is exactly where price likes to wick before it turns. Sitting a buffer past the obvious line keeps you from being the liquidity a stop-hunt feeds on. Treat "the market respects structure" as a strong tendency, not a law. Levels break and stop-hunts happen, so a structural stop is a high-probability statement of invalidation, not a guarantee.
If price trades through the level that defined your thesis, the setup failed. You were not unlucky, you were wrong.
ATR and volatility stops
ATR (Average True Range) measures the average size of a bar's range over a lookback, most often 14 periods. It is a volatility number, not a direction number, and it answers one question: how far does this instrument routinely move on this timeframe? Once you know that, you can scale the stop to current volatility instead of a fixed distance. The formula is direct:
- Stop distance = ATR x multiplier
- Long stop = entry - (ATR x multiplier)
- Short stop = entry + (ATR x multiplier)
The multipliers traders reach for are conventions, not laws. Intraday and day-trading timeframes commonly use roughly 1.5x to 2x ATR. Swing traders often run about 2x to 3x. Longer-horizon position traders sometimes go 3x to 4x. None of these is a proven optimum, and different instruments, timeframes, and strategies justify different numbers. A volatility-scaled stop earns its keep by widening when the market is volatile and tightening when it is calm, so a normal wiggle does not eject you while a real adverse move does.
One rule is non-negotiable. Use the ATR of the same timeframe you trade. A 14-period ATR on a 1-minute chart and a 14-period ATR on a daily chart describe completely different distances, and mixing them hands you a stop that is either absurdly tight or absurdly wide for what you are actually doing. The 14-period lookback is a standard default, not a requirement. Other lookbacks are fine if you know why you picked them.
| Style | Common ATR multiple | Character |
|---|---|---|
| Intraday / day | ~1.5x to 2x | Tighter, tracks fast noise |
| Swing | ~2x to 3x | Room for overnight range |
| Position | ~3x to 4x | Ignores daily chop |
Time stops
A time stop exits a trade that has not done its job within a preset window: a number of bars, a set number of minutes, or a session boundary. It is not a price stop and it does not replace one. Keep the price stop for catastrophe protection. The time stop handles a different failure mode, the "right idea, dead timing" case where price just sits there and does nothing.
Two reasons justify it. One is opportunity cost. Capital and attention sit locked in a trade going nowhere while other setups walk past. The other is thesis decay. Many entries are time-sensitive, and the edge behind them expires. If the breakout has not broken out, or the reaction to a level never showed, the reason to hold is quietly evaporating even with the price stop untouched. A time stop makes that expiry explicit instead of leaving you to rationalize a stale position.
Decide the window before you enter, not while you stare at a flat trade willing it to move. Write down the bar count or the clock, and treat it as part of the plan.
The dollar-stop trap
Now the single most common beginner mistake, stated plainly. A trader decides "I'll risk $150," and drops the stop $150 from entry no matter what the chart says. The stop lands wherever the wallet chose, not where structure or volatility did. If that dollar distance happens to fall inside the instrument's normal noise, you have guaranteed random stop-outs. Ordinary wiggle takes you out, then price does exactly what you expected without you. The market never knew where your $150 was.
The fix is a reordering, not a new technique. The dollar figure belongs in the sizing step, never in the stop-placement step. Find the stop from the chart first, and only then does money enter the conversation. Anyone building a repeatable process around fixed risk should read this next to risk-reward and R-multiples and a proper approach to position sizing on funded accounts, because the dollar-stop trap is really a sequencing failure between those two ideas.
A dollar figure in the stop-placement step drops your exit into a spot the market never cared about. Move the dollars to the sizing step.
Stop first, then size
The order of operations is fixed. Find the technical stop from the chart, then work out how many contracts fit your fixed risk. Never size first and jam the stop to fit the size. The relationship is a single identity:
- Stop distance (points) x Dollars per point x Number of contracts = Fixed dollar risk
- Solve for size: Contracts = Fixed dollar risk / (Stop distance in points x Dollars per point)
- Round DOWN to a whole contract. You cannot trade a fraction, and rounding up overshoots your risk.
Work a full example. You are long the E-mini S&P 500 (ES) with fixed risk of $300 per trade. Entry is 6,800.00. The nearest swing low that defines the long is 6,793.00, and you place the stop 1.00 point below it to sit behind resting liquidity, so the stop price is 6,792.00. Stop distance is 6,800.00 minus 6,792.00, which is 8.00 points. ES is worth $50 per point, so one contract risks 8.00 x $50 = $400. Solve for size: $300 / $400 = 0.75 contracts. You cannot trade 0.75 ES, and rounding down gives zero full contracts. A single ES would risk $400, which blows past the $300 budget, so ES does not fit this trade.
The funded-account move is to change the instrument, not the stop. Switch to MES, the Micro E-mini S&P, worth $5 per point, and keep the same 8-point structural stop. One MES now risks 8.00 x $5 = $40. Size is $300 / $40 = 7.5, which rounds down to 7 MES. Actual risk taken is 7 x $40 = $280, sitting under the $300 budget with $20 of headroom. Check the identity: 8.00 x $5 x 7 = $280. Confirmed.
Now watch the trap in numbers. Say you decide instead "I want 3 ES, and I'll risk $300," so you set the stop at $300 / (3 x $50) = 2.00 points away, at 6,798.00. That 2-point stop sits far inside the real structure, because the actual invalidation is 8 points away at 6,792.00. A 2-point stop on ES is inside routine noise, so you get shaken out on a wiggle that never touched your thesis. You sized first and jammed the stop to fit, which is precisely the error to avoid.
| Contract | $ per point | Tick (0.25 pt) | Risk on 8-pt stop |
|---|---|---|---|
| ES | $50 | $12.50 | $400 |
| NQ | $20 | $5.00 | $160 |
| MES | $5 | $1.25 | $40 |
| MNQ | $2 | $0.50 | $16 |
MGC (Micro Gold) runs $10 per point with a $1.00 tick at 0.10, listed here so you can see how much a smaller multiplier bends the arithmetic. These are current CME standard specs. Always confirm the live spec and tick value for the exact contract and month you trade before sizing real money, because the whole calculation rides on them.
Stops on a funded account
On a funded or prop account, the stop has to respect the drawdown limit, trailing or end-of-day, not just a per-trade R figure. Stop distance and buffer together set the largest size you are allowed to carry. This is the direct link between stop placement, R-multiples, and position sizing. The stop defines 1R, the remaining buffer caps total R exposure, and size is the free variable you adjust. The instrument is the other lever.
When the correct structural stop makes the position too large for your remaining buffer, the answer is never to tighten the stop into the noise. There are exactly two valid moves. You can pass the trade. Or you can trade a smaller-multiplier instrument (MES instead of ES, MNQ instead of NQ) so the same structural distance costs fewer dollars per contract and fits the buffer. Both are honest. Jamming the stop tighter to force the trade is not.
The exact drawdown mechanics, trailing versus end-of-day and the precise reset behavior, vary by firm. Verify the specific firm's rules before you use them to size, and do not assume a generic model. Here is the part most traders miss: the account rules reward survival, not bravado. Passing a trade you cannot size to its true stop is a skill, not a failure. "No trade" is a valid position.
Risk lives in size, not in stop distance. Most traders tighten the stop when they should shrink the position.
This is also where the honest tradeoff lives. A stop is a request, not a guarantee. A stop-market order becomes a market order the moment it is touched and fills at the next available price, so in fast or gapping markets that fill can land worse than your level. That gap is slippage, and it is real risk you cannot fully design away. A stop-limit order bounds the fill price but can fail to fill at all in a fast move, leaving you stuck in a position you meant to exit. You pick price certainty or execution certainty. You cannot erase both. If you want the mechanics of each order type, the guide to futures order types covers bracket, OCO, stop, and trailing behavior in detail.
Moving stops the right way
Two stop moves are allowed. You may move to breakeven once the trade has proven itself, and you may trail behind newly formed structure, behind each higher swing low in an uptrend and each lower swing high in a downtrend. Both lock in progress without inventing a new thesis. They follow the chart, which is the same discipline that placed the stop in the first place.
One move is never allowed: widening a stop to dodge a loss. Widening corrupts your R, because your realized loss is now bigger than the 1R you planned and sized for. It is the classic tell of a losing habit, trading to avoid the feeling of losing rather than to protect capital. The urge to widen is the same urge that makes people place stops for comfort in the first place. That is why the rule is absolute. If you feel the pull to widen, that pull is the signal to close, not to move.
Widening a stop turns one clean 1R loss into an oversized one. If you want to move it out, close instead.
Run the whole thing as a checklist before every entry. Mark where the trade is objectively wrong and add a small buffer beyond it. Sanity-check that distance against ATR: if the structural stop is tighter than about 1.5x ATR of your timeframe, it is inside the noise, so use the ATR distance or skip. Lock the stop and do not touch it to make sizing convenient. Then size: contracts equal fixed risk divided by (stop distance in points times dollars per point), rounded down. Confirm the size fits your remaining drawdown buffer. If it does not, pass the trade or drop to a micro instrument, and never tighten the stop to fit. Decide the time-stop window now and write it down. Choose stop-market or stop-limit with eyes open. And hold the management rule: move to breakeven or trail behind new structure, never widen.
Two places this whole approach is not the answer. Structural and ATR stops assume the instrument is liquid enough to have meaningful structure and a stable ATR. In thin, illiquid, or news-shocked conditions, structure gets unreliable and ATR lags a volatility spike, so the honest move is smaller size or standing aside, not a cleverer stop. And a trade copier fixes none of this. A copier such as Thor replicates entries, stops, and exits across accounts at low latency, which is genuinely useful when your process is sound, but it does not improve where the stop sits. If the source stop is placed for comfort or dropped into noise, the copier faithfully multiplies that mistake across every account, and slippage can run worse on followers. A copier scales a good process. It launders nothing.
Frequently asked questions
What are the two jobs of a stop-loss?
A stop does two separate jobs at once. First it invalidates the trade idea by marking the price where your thesis is objectively proven wrong, and second it caps the dollar loss if you are wrong. These are different tasks: the invalidation level is a fact about the chart, while the dollar loss is a fact about your account that you control through position sizing. A good stop handles invalidation, and sizing handles the loss cap.
How do I calculate position size from a stop-loss in futures?
Use the identity that stop distance in points times dollars per point times number of contracts equals your fixed dollar risk. Rearranged, contracts equal fixed dollar risk divided by (stop distance in points times dollars per point), and you always round down to a whole contract. For example, a $300 risk with an 8-point stop on ES at $50 per point gives $300 divided by $400, which is 0.75, rounding down to zero, so ES does not fit and you switch to a micro like MES.
What is an ATR stop and what multiplier should I use?
An ATR stop scales the stop distance to volatility using stop distance equals ATR times a multiplier, where ATR is the Average True Range over a lookback, most commonly 14 periods. Common conventions are roughly 1.5x to 2x ATR for intraday trading, 2x to 3x for swing trades, and 3x to 4x for position trades, but these are conventions rather than proven optimums. Always use the ATR of the same timeframe you trade, since a 14-period ATR on a 1-minute chart is a completely different distance than on a daily chart.
What is the dollar-stop trap?
The dollar-stop trap is placing your stop at a fixed dollar loss, such as $150 away, regardless of what the chart says. This drops the stop into a location chosen by your wallet rather than by structure or volatility, and if that distance falls inside the instrument's normal noise you guarantee random stop-outs on ordinary wiggle. The dollar figure belongs in the sizing step, never in the stop-placement step, which is why this is the single most common beginner mistake.
Should I ever tighten my stop to fit a funded account's drawdown limit?
No. If the correct structural stop makes the position too large for your remaining drawdown buffer, tightening the stop into the noise only guarantees random stop-outs. The two valid moves are to pass the trade entirely, or to trade a smaller-multiplier instrument like MES instead of ES so the same structural distance costs fewer dollars per contract. Drawdown mechanics vary by firm, so verify the specific trailing or end-of-day rules before sizing.
Can I widen a stop-loss once I am in a trade?
You should never widen a stop to avoid taking a loss. Widening corrupts your R because your realized loss becomes bigger than the 1R you planned and sized for, and it is the classic tell of trading to avoid the feeling of losing rather than to protect capital. The only allowed moves are to breakeven once the trade proves itself and trailing behind newly formed structure. If you feel the urge to widen, treat it as a signal to close, not to move the stop.
Does a trade copier improve where my stop is placed?
No. A trade copier replicates entries, stops, and exits across accounts, but it does not improve where the stop sits on the chart. If the source stop is placed for comfort or dropped into noise, the copier faithfully multiplies that bad stop across every account, and slippage can run worse on followers because of the extra hop and staggered fills. A copier scales a good process, but it cannot rescue a stop that is wrong or a size that is too large.
What is a time stop and how is it different from a price stop?
A time stop exits a trade that has not done its job within a preset window, measured in bars, minutes, or a session boundary. It addresses opportunity cost and thesis decay, the case where the edge was time-sensitive and the move never started. It does not replace the price stop; you keep the price stop for catastrophe protection while the time stop handles the right-idea, dead-timing situation.